Подпишитесь на рассылки о научных публикациях
4 year student, Oksana Anatoliivna Kychko
NTUU "KPI", Department of Linguistics, Kyiv
One of the foundations of modern linguistics is the maxim of categoricity: language is categorical. Numbers play no role, or, where they do, they are artifacts of nonlinguistic performance factors. Thus, while it is widely recognized that real language can be highly variable, gradient, and rich in continua, many linguists would argue that the competence that underlies such ‘‘performance factors'' consists of well-defined discrete categories and categorical grammaticality criteria. Performance may be full of fuzziness, gradience, and continua, but linguistic competence is not.
However, a groundswell of recent results challenge the idea that linguistic competence is categorical and discrete. While linguistic phenomena such as phonological and morphological alternations and syntactic well-formedness judgments tend to be modeled as categorical, it has become increasingly clear that alternations and judgments display properties of continua and show markedly gradient behavior. Moreover, psycholinguistic experiments demonstrate that speakers' well-formed judgments of words and sentences are extremely well predicted by the combined probabilities of their subparts.
While generative approaches to linguistics have evolved to capture the endpoints of such distributions, there is growing interest in the relatively unexplored gradient middle ground, and a growing realization that concentrating on the extremes of continua leaves half the phenomena unexplored and unexplained. On the contrary, a probabilistic approach can push the boundaries of linguistic theory forward, by substantially enriching the current state of knowledge. Probabilistic linguistics increases the range of data for which a theory can account, and for which it must be accountable.
In recent years, a strong consensus has emerged that human cognition is based on probabilistic processing. However, the fact that probability theory is an increasingly useful and important tool in cognitive science does not make it automatically suitable for modeling language. To be convinced of its suitability, readers should rightly demand evidence that the language faculty itself displays probabilistic properties. A briefly outlined nature of this evidence is given below.
Language changes over time-a process that is usually echoed synchronically across age groups. Zuraw  provides evidence that language change can result from probabilistic inference on the part of listeners, and she argues that probabilistic reasoning ‘‘could explain the maintenance of lexical regularities over historical time''. It is well accepted that language does not just vary across time-it is inherently variable. There is no known case, for example, where analogous phonemes have exactly the same implementation across two languages (Pierrehumbert ).
Acquiring a language or dialect, then, involves not just identifying its phonemes, but also learning the extremely subtle patterns of production and allophony relevant to each phoneme in that language. Within a particular language, production patterns differ across individuals, depending on aspects of identity (Mendoza-Denton, Hay, and Jannedy ).
Within individuals, production patterns differ on the basis of stylistic factors such as addressee, context, and topic, and this stylistic variation to a large degree echoes the variation present across members of society. Knowledge of variation, then, must form part of linguistic competence, since individuals can manipulate their implementation of phonetic variants to portray linguistic and extralinguistic information. And individuals differ not only in the specific variants they use in different contexts, but also in the frequency with which they use them. Knowledge of variation must involve knowledge of frequencies (Mendoza-Denton, Hay, and Jannedy ).
One striking clue to the importance of probabilities in language comes from the wealth of frequency effects that pervade language representation, processing, and language change. Frequent words are recognized faster than infrequent words, and there is a bias toward interpreting ambiguous words in terms of their more frequent meanings (Jurafsky ). Frequent words lead leniting changes (Zuraw ) and are more prone to reduction in speech (Jurafsky ; Mendoza-Denton, Hay, and Jannedy ). Frequent combinations of phonemes (Pierrehumbert ) and structures (Manning) are perceived as more grammatical, or well formed, than infrequent combinations.
The relative frequency of derived words and their bases affects the morphological decomposability of complex words (Baayen ). These are many frequency effects that influence language perception, production, and representation. Frequency affects language processes, and so it must be represented somewhere. The language-processing system tracks, records, and exploits frequencies of various kinds of events.
We can best model many of these effects by making explicit the link between frequency and probability. Probability theory provides well-articulated methods for modeling frequency, and it provides researchers with the tools to work not only with the frequency of events, but also with the frequency of combinations of events. One can thus estimate the probability of complex events (such as sentences) by combining the probabilities of their subparts.
The presence of frequency effects is not in itself sufficient to warrant adopting a probabilistic view. It is conceivable that at least some of the frequency effects could occur without any kind of probabilistic effect. However, the presence of frequency effects does provide evidence that the basic building blocks of probability theory are stored and exploited. Just as the complete absence of frequency effects would challenge the foundations of probabilistic linguistics, so their overwhelming presence adds weight to the claim that the language faculty is inherently probabilistic.
Language displays all the hallmarks of a probabilistic system. Categories and well-formedness are gradient, and frequency effects are everywhere. We believe all evidence points to a probabilistic language faculty. Knowledge of language should be understood not as a minimal set of categorical rules or constraints, but as a (possibly redundant) set of gradient rules, which may be characterized by a statistical distribution.
•1. Baayen, R. H., and Schreuder, R., Towards a psycholinguistic computational model for morphological parsing. Philosophical Transactions of the Royal Society (Series A: Mathematical, Physical and Engineering Sciences) 358, pp.1-13, 2000.
•2. Hay, J. B., Mendoza-Denton, N., and Jannedy, S., Style and economy: When frequency effects collide. Paper presented at the 7th New Zealand Language and Society Conference, Auckland, New Zealand, 700 p., 2000.
•3. Jurafsky, D., A probabilistic model of lexical and syntactic access and disambiguation. Cognitive Science 20, pp. 137-194, 1996.
•4. Pierrehumbert, J., Exemplar dynamics: Word frequency, lenition and contrast. In J. Bybee and P. Hopper, eds., Frequency and the emergence of linguistic structure. Amsterdam: John Benjamins, pp. 137-157, 2001.
•5. Zuraw, K., Patterned exceptions in phonology. Ph.D. thesis, University of California, Los Angeles, p.500, 2000.